People have been selectively breeding animals and plants for thousands of years – to provide better sources of materials for food and clothing or for decoration.

Carassus auratus is a small brown fish related to the carp. But, occasionally, an individual appears with an unusual shape or colour (usually red or yellow). By selectively breeding these unusual individuals with each other, it was possible to develop named varieties of what we call “goldfish”. Some examples are shown in the picture above. Goldfish breeding appears to have started in China over a thousand years ago.

Selective breeding was first investigated scientifically by Gregor Mendel (1822-1884) a monk (but also a biologist and mathematician) who lived in the Austrian Empire in a country that is now the Czech Republic. He did experiments on breeding the pea plant Pisum sativum. In this post I am going to concentrate on his results on the colour of the flowers of this plant. He made many other observations but flower colour enables us to understand what he found without looking at all his other results. The plant can have purple or white flowers, as shown in the picture above. If they cannot pollinate each other, purple-flowered plants produce only purple-flowered offspring; white-flowered plants produce only white-flowered offspring.

Mendel artificially pollinated parent plants with different colour flowers – the resulting plants, the F1 generation, all had purple flowers.

He then used the F1 generation as parents for the F2 generation; in this generation 75% of the plants had purple flowers but 25% had white flowers.

Mendel explained this result by supposing that the plants could carry information to produce either purple or white flowers – we now call the source of this information a gene. He further supposed that there were two forms of the gene for flower colour – we now call the different forms of a gene its alleles. Suppose purple-flowered plants that consistently produce purple-flowered offspring contain two identical alleles, that I’ve denoted by C in the picture. Suppose also that white-flowered plants that consistently produce white-flowered offspring contain two identical alleles, that I’ve denoted by c in the picture. When a purple flower pollinates a white flower, the new plants must have a pair of non-identical alleles, denoted by Cc. If the C alleles control the resulting flower colour, these new plants will be purple. We then say that the C allele is dominant and the c allele is recessive. Notice that purple flowers all look the same (we say they have the same phenotype) but can have different genetic compositions, CC or Cc (we say they have different genotypes). Its very important not to confuse genotype and phenotype – if we do, we can easily make false assumptions.

To test Mendel’s model, let’s look at what happens in the F2 generation. A new plant can receive a C allele from the first parent and a C allele from the second – it will have purple flowers. It can receive a C allele from the first parent and a c allele from the second – it will have purple flowers. Or it can receive a c allele from the first parent and a C allele from the second – it will also have purple flowers. But it can receive a c allele from the first parent and a c allele from the second – it will then have white flowers. So, 75% of the F2 generation have purple flowers and 25% will have white flowers. Mendel’s model works.

If you are a gardener, you will now realise why (1) F1 hybrid seeds are expensive and (2) why it’s not a good idea to save the seeds of the resulting plants for next year.

Mendel was lucky in that the colours of his flowers depended on only one gene. Most of our characteristics are controlled by more than one gene and so understanding how we inherit these characteristics is usually much more complicated than understanding the colours of pea flowers. Even something as simple as human eye colour depends on several different genes. So, if you read an article in a newspaper or on the web, about a gene for intelligence, or some other human characteristic, it is almost certainly an oversimplification. Also our genes provide the information for something – they don’t guarantee that this information will be implemented. What happens also depends on environmental factors. For example, if the flower buds of a pea plant are killed by frost, it won’t have the flowers predicted by its genotype.

Follow-up posts

21.9 Chromosomes 1
21.10 Chromosomes 2
21.11 Genes are made of DNA